coreylowman dfdx issues

coreylowman / dfdx

Deep learning in Rust, with shape checked tensors and neural networks

Other

1.71k stars 99 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Safetensor panics instead of returning Err

#928 opfromthestart opened 1 month ago
0
Make tests pass on stable + support Pool2D & Conv{1,2}d

#927 twitchyliquid64 opened 2 months ago
0
Tests failing on HEAD /w cargo 1.78.0

#926 twitchyliquid64 opened 2 months ago
0
Kernels written in rust-gpu

#925 LegNeato opened 5 months ago
1
can not perform tensor operations (add/sub/mul/...) on tensors with non float datatype

#924 lucomotive opened 5 months ago
0
fix: small typo in advanced-train-loop.rs

#923 LeoDog896 opened 7 months ago
0
docs(readme): fix examples hyperlink

#922 mrtnvgr opened 7 months ago
0
Non-trainable parameters?

#921 hovinen opened 7 months ago
2
Avoid CI errors

#920 swfsql opened 7 months ago
0
Added `TryUnstack` for tensors.

#919 swfsql opened 8 months ago
0
Add Mamba (minimal)

#918 swfsql closed 7 months ago
1
Add split_tensor_along method

#917 swfsql opened 8 months ago
0
Add RMS normalization

#916 swfsql opened 8 months ago
0
Add SiLU activation function

#915 swfsql opened 8 months ago
0
Impl `core::ops::Sub` for Dim types

#914 swfsql opened 8 months ago
0
Safetensors changes

#913 swfsql opened 8 months ago
0
Remove functions calling the deprecated `_MM_SET_FLUSH_ZERO_MODE`

#912 rainiwu opened 8 months ago
3
make non-const conv1d available to stable rust

#911 optman closed 8 months ago
0
CUDA kernels missing __hmin and __hmax

#910 opfromthestart opened 8 months ago
1
Add missing backtick (`) in crate-level documentation

#909 rainiwu closed 8 months ago
0
OpenXLA Support

#908 vyeevani opened 8 months ago
3
Unable to build with old CUDA version (`CUDA_COMPUTE_CAP = 52`)

#907 nikvoid opened 8 months ago
0
feat(wgpu): add to_dtype kernel

#906 DonIsaac closed 8 months ago
0
How does one update one model from another model?

#905 hovinen closed 7 months ago
1
feat(webgpu): add tensor operations

#904 DonIsaac opened 9 months ago
1
Add `nn::AdaptiveAvgPool2D`

#903 quietlychris opened 9 months ago
0
Gradient Clipping

#902 swfsql opened 9 months ago
0
small doc typo fix

#901 HyperCodec closed 9 months ago
0
Unclear how to handle error type in `dfdx::nn::LoadFromNpz::load`

#900 hovinen opened 9 months ago
0
Run tests with miri

#899 swfsql closed 7 months ago
1
Reduce test sizes

#898 swfsql closed 7 months ago
1
Run tests with miri

#897 swfsql closed 7 months ago
1
Implement AbsKernelOp for WebGPU backend

#896 favilo closed 9 months ago
7
Add `Prodigy` optimizer

#895 swfsql closed 7 months ago
1
Add `Prodigy` optimizer

#894 swfsql closed 7 months ago
1
Implement Webgpu device

#893 favilo closed 10 months ago
4
Split `TryConcatAlong` into different traits

#892 swfsql closed 10 months ago
3
Split `TryConcatAlong` into different traits

#891 swfsql closed 10 months ago
0
Adds `OUTPUT_PADDING` to `ConvTrans2D`

#890 swfsql opened 10 months ago
0
Add `OUTPUT_PADDING` to `ConvTrans2D`

#889 swfsql opened 10 months ago
0
Send/Sync for Device

#888 blogle opened 10 months ago
1
On older GPU there are no hcos/hsin/hlog/hexp

#887 optman closed 10 months ago
1
Replace explicit features and paths on generated code

#886 swfsql closed 10 months ago
0
Replace explicit features and paths on generated code

#885 swfsql closed 10 months ago
0
Add `forward_mut` in `Sequential` derive

#884 swfsql closed 11 months ago
0
Input Wrapper

#883 swfsql closed 7 months ago
2
make device capable of stacking and concatenating usize type

#882 optman closed 11 months ago
0
Bug: `Sequential` macro provide `forward_mut` as `forward`

#881 uselessgoddess closed 11 months ago
0
Examples or resources for autodiff with 2 networks?

#880 Cobrand opened 11 months ago
0
Question / clarification regarding heap allocations

#879 emchristiansen closed 11 months ago
2